Eubacteria export numerous proteins across the plasma membrane into either the periplasmic space (Gram-negative species), or the growth medium (Gram-positive species). The Gram-positive eubacterium Bacillus subtilis and, in particular, its close relatives Bacillus amyloliquefaciens and Bacillus licheniformis are well known for their high capacity to secrete proteins (at gram per liter concentrations) into the medium. This property, which allows the efficient separation of (secreted) proteins from the bulk cytoplasmic protein complement, has led to the commercial exploitation of the latter bacilli as important “cell factories.” Despite their high capacity to secrete proteins of Gram-positive origin, the secretion of recombinant proteins from Gram-negative eubacterial or eukaryotic origin by Bacillus species is often inefficient. This can be due to a variety of (potential) bottlenecks in the secretion pathway, such as poor targeting to the membrane, pre-translocational folding, inefficient translocation, slow or incorrect post-translocational folding of the secretory protein, and proteolysis. Notably, many of these problems relate to the specific properties of the general secretory (Sec) pathway that was, so far, used in all documented attempts to apply bacilli for the secretion of heterologous proteins of commercial or biomedical value.
General strategies for the secretion of heterologous proteins by bacilli are based on the in-frame fusion of the respective protein with an amino-terminal signal peptide that directs this protein into the Sec-dependent secretion pathway. Upon translocation across the membrane, the signal peptide is removed by a signal peptidase, which is a prerequisite for the release of the translocated protein from the membrane, and its secretion into the medium. As exemplified with human interleukin-3, which is secreted by B. licheniformis at gram per liter concentrations, this strategy allows protein production at commercially significant levels.
Two major hurdles have been identified for the secretion of heterologous proteins via the Sec-dependent route. The first one is the translocation process by the Sec machinery, which is composed of a proteinaceous channel in the membrane (consisting of SecY, SecE, SecG and SecDF-YrbF) and a translocation motor (SecA). The Sec machinery is known to ‘thread’ its substrates in an unfolded state through the membrane. Consequently, this machinery is inherently incapable of translocating proteins that fold in the cytosol. A second bottleneck has been identified for other heterologous proteins that are translocated correctly but fold slowly or incorrectly in the cell wall environment, probably because this compartment lacks the appropriate chaperone molecules to assist in their folding. Molecular chaperones of the Hsp60 and Hsp70 classes are essential for the folding of many proteins, but these are all absent from bacterial extracytoplasmic compartments. As the membrane-cell wall environment of bacilli is highly proteolytic, slowly or incorrectly folding translocated proteins are often degraded before being secreted into the medium. Consequently, protein secretion via the Sec pathway is a highly efficient tool for the production of only a subset of heterologous proteins.
Protein production and secretion from Bacillus species is a major production tool with a market of over $1 billion per year. However, as noted above, the standard export technologies, based on the well-characterized general secretory (Sec) pathway, are frequently inapplicable for the production of proteins. Thus, it would be beneficial to have an alternative mechanism for the production and secretion proteins.